ice: Handle critical FW error during admin queue initialization

A race condition between FW and SW can occur between admin queue setup and
the first command sent. A link event may occur and FW attempts to notify a
non-existent queue. FW will set the critical error bit and disable the
queue. When this happens retry queue setup.

Signed-off-by: Evan Swanson <evan.swanson@intel.com>
Signed-off-by: Anirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This commit is contained in:
Evan Swanson 2020-05-15 17:42:14 -07:00 committed by Jeff Kirsher
parent 1960827570
commit b5c7f857e5
5 changed files with 95 additions and 69 deletions

View File

@ -12,6 +12,7 @@ do { \
(qinfo)->sq.bal = prefix##_ATQBAL; \
(qinfo)->sq.len_mask = prefix##_ATQLEN_ATQLEN_M; \
(qinfo)->sq.len_ena_mask = prefix##_ATQLEN_ATQENABLE_M; \
(qinfo)->sq.len_crit_mask = prefix##_ATQLEN_ATQCRIT_M; \
(qinfo)->sq.head_mask = prefix##_ATQH_ATQH_M; \
(qinfo)->rq.head = prefix##_ARQH; \
(qinfo)->rq.tail = prefix##_ARQT; \
@ -20,6 +21,7 @@ do { \
(qinfo)->rq.bal = prefix##_ARQBAL; \
(qinfo)->rq.len_mask = prefix##_ARQLEN_ARQLEN_M; \
(qinfo)->rq.len_ena_mask = prefix##_ARQLEN_ARQENABLE_M; \
(qinfo)->rq.len_crit_mask = prefix##_ARQLEN_ARQCRIT_M; \
(qinfo)->rq.head_mask = prefix##_ARQH_ARQH_M; \
} while (0)
@ -641,72 +643,6 @@ init_ctrlq_free_sq:
return ret_code;
}
/**
* ice_init_all_ctrlq - main initialization routine for all control queues
* @hw: pointer to the hardware structure
*
* Prior to calling this function, the driver MUST* set the following fields
* in the cq->structure for all control queues:
* - cq->num_sq_entries
* - cq->num_rq_entries
* - cq->rq_buf_size
* - cq->sq_buf_size
*
* NOTE: this function does not initialize the controlq locks.
*/
enum ice_status ice_init_all_ctrlq(struct ice_hw *hw)
{
enum ice_status ret_code;
/* Init FW admin queue */
ret_code = ice_init_ctrlq(hw, ICE_CTL_Q_ADMIN);
if (ret_code)
return ret_code;
ret_code = ice_init_check_adminq(hw);
if (ret_code)
return ret_code;
/* Init Mailbox queue */
return ice_init_ctrlq(hw, ICE_CTL_Q_MAILBOX);
}
/**
* ice_init_ctrlq_locks - Initialize locks for a control queue
* @cq: pointer to the control queue
*
* Initializes the send and receive queue locks for a given control queue.
*/
static void ice_init_ctrlq_locks(struct ice_ctl_q_info *cq)
{
mutex_init(&cq->sq_lock);
mutex_init(&cq->rq_lock);
}
/**
* ice_create_all_ctrlq - main initialization routine for all control queues
* @hw: pointer to the hardware structure
*
* Prior to calling this function, the driver *MUST* set the following fields
* in the cq->structure for all control queues:
* - cq->num_sq_entries
* - cq->num_rq_entries
* - cq->rq_buf_size
* - cq->sq_buf_size
*
* This function creates all the control queue locks and then calls
* ice_init_all_ctrlq. It should be called once during driver load. If the
* driver needs to re-initialize control queues at run time it should call
* ice_init_all_ctrlq instead.
*/
enum ice_status ice_create_all_ctrlq(struct ice_hw *hw)
{
ice_init_ctrlq_locks(&hw->adminq);
ice_init_ctrlq_locks(&hw->mailboxq);
return ice_init_all_ctrlq(hw);
}
/**
* ice_shutdown_ctrlq - shutdown routine for any control queue
* @hw: pointer to the hardware structure
@ -751,6 +687,82 @@ void ice_shutdown_all_ctrlq(struct ice_hw *hw)
ice_shutdown_ctrlq(hw, ICE_CTL_Q_MAILBOX);
}
/**
* ice_init_all_ctrlq - main initialization routine for all control queues
* @hw: pointer to the hardware structure
*
* Prior to calling this function, the driver MUST* set the following fields
* in the cq->structure for all control queues:
* - cq->num_sq_entries
* - cq->num_rq_entries
* - cq->rq_buf_size
* - cq->sq_buf_size
*
* NOTE: this function does not initialize the controlq locks.
*/
enum ice_status ice_init_all_ctrlq(struct ice_hw *hw)
{
enum ice_status status;
u32 retry = 0;
/* Init FW admin queue */
do {
status = ice_init_ctrlq(hw, ICE_CTL_Q_ADMIN);
if (status)
return status;
status = ice_init_check_adminq(hw);
if (status != ICE_ERR_AQ_FW_CRITICAL)
break;
ice_debug(hw, ICE_DBG_AQ_MSG,
"Retry Admin Queue init due to FW critical error\n");
ice_shutdown_ctrlq(hw, ICE_CTL_Q_ADMIN);
msleep(ICE_CTL_Q_ADMIN_INIT_MSEC);
} while (retry++ < ICE_CTL_Q_ADMIN_INIT_TIMEOUT);
if (status)
return status;
/* Init Mailbox queue */
return ice_init_ctrlq(hw, ICE_CTL_Q_MAILBOX);
}
/**
* ice_init_ctrlq_locks - Initialize locks for a control queue
* @cq: pointer to the control queue
*
* Initializes the send and receive queue locks for a given control queue.
*/
static void ice_init_ctrlq_locks(struct ice_ctl_q_info *cq)
{
mutex_init(&cq->sq_lock);
mutex_init(&cq->rq_lock);
}
/**
* ice_create_all_ctrlq - main initialization routine for all control queues
* @hw: pointer to the hardware structure
*
* Prior to calling this function, the driver *MUST* set the following fields
* in the cq->structure for all control queues:
* - cq->num_sq_entries
* - cq->num_rq_entries
* - cq->rq_buf_size
* - cq->sq_buf_size
*
* This function creates all the control queue locks and then calls
* ice_init_all_ctrlq. It should be called once during driver load. If the
* driver needs to re-initialize control queues at run time it should call
* ice_init_all_ctrlq instead.
*/
enum ice_status ice_create_all_ctrlq(struct ice_hw *hw)
{
ice_init_ctrlq_locks(&hw->adminq);
ice_init_ctrlq_locks(&hw->mailboxq);
return ice_init_all_ctrlq(hw);
}
/**
* ice_destroy_ctrlq_locks - Destroy locks for a control queue
* @cq: pointer to the control queue
@ -1049,9 +1061,15 @@ ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq,
/* update the error if time out occurred */
if (!cmd_completed) {
ice_debug(hw, ICE_DBG_AQ_MSG,
"Control Send Queue Writeback timeout.\n");
status = ICE_ERR_AQ_TIMEOUT;
if (rd32(hw, cq->rq.len) & cq->rq.len_crit_mask ||
rd32(hw, cq->sq.len) & cq->sq.len_crit_mask) {
ice_debug(hw, ICE_DBG_AQ_MSG, "Critical FW error.\n");
status = ICE_ERR_AQ_FW_CRITICAL;
} else {
ice_debug(hw, ICE_DBG_AQ_MSG,
"Control Send Queue Writeback timeout.\n");
status = ICE_ERR_AQ_TIMEOUT;
}
}
sq_send_command_error:

View File

@ -34,6 +34,8 @@ enum ice_ctl_q {
/* Control Queue timeout settings - max delay 250ms */
#define ICE_CTL_Q_SQ_CMD_TIMEOUT 2500 /* Count 2500 times */
#define ICE_CTL_Q_SQ_CMD_USEC 100 /* Check every 100usec */
#define ICE_CTL_Q_ADMIN_INIT_TIMEOUT 10 /* Count 10 times */
#define ICE_CTL_Q_ADMIN_INIT_MSEC 100 /* Check every 100msec */
struct ice_ctl_q_ring {
void *dma_head; /* Virtual address to DMA head */
@ -59,6 +61,7 @@ struct ice_ctl_q_ring {
u32 bal;
u32 len_mask;
u32 len_ena_mask;
u32 len_crit_mask;
u32 head_mask;
};

View File

@ -39,6 +39,7 @@
#define PF_MBX_ARQH_ARQH_M ICE_M(0x3FF, 0)
#define PF_MBX_ARQLEN 0x0022E480
#define PF_MBX_ARQLEN_ARQLEN_M ICE_M(0x3FF, 0)
#define PF_MBX_ARQLEN_ARQCRIT_M BIT(30)
#define PF_MBX_ARQLEN_ARQENABLE_M BIT(31)
#define PF_MBX_ARQT 0x0022E580
#define PF_MBX_ATQBAH 0x0022E180
@ -47,6 +48,7 @@
#define PF_MBX_ATQH_ATQH_M ICE_M(0x3FF, 0)
#define PF_MBX_ATQLEN 0x0022E200
#define PF_MBX_ATQLEN_ATQLEN_M ICE_M(0x3FF, 0)
#define PF_MBX_ATQLEN_ATQCRIT_M BIT(30)
#define PF_MBX_ATQLEN_ATQENABLE_M BIT(31)
#define PF_MBX_ATQT 0x0022E300
#define PRTDCB_GENC 0x00083000

View File

@ -5207,6 +5207,8 @@ const char *ice_stat_str(enum ice_status stat_err)
return "ICE_ERR_AQ_NO_WORK";
case ICE_ERR_AQ_EMPTY:
return "ICE_ERR_AQ_EMPTY";
case ICE_ERR_AQ_FW_CRITICAL:
return "ICE_ERR_AQ_FW_CRITICAL";
}
return "ICE_ERR_UNKNOWN";

View File

@ -37,6 +37,7 @@ enum ice_status {
ICE_ERR_AQ_FULL = -102,
ICE_ERR_AQ_NO_WORK = -103,
ICE_ERR_AQ_EMPTY = -104,
ICE_ERR_AQ_FW_CRITICAL = -105,
};
#endif /* _ICE_STATUS_H_ */